The invention relates to a method of detecting a lack of driver activity at the steering wheel of a motor vehicle, wherein the moment imparted by hand to the steering wheel by the driver is repeatedly measured, and in the event that in several successive measurements the amount of the magnitude of the moment imparted by hand does not exceed a threshold value, a lack of driver activity is inferred. With respect to the state of the art, reference is made, in particular, to German Patent document DE 10 2008 021 150 A1, and further in the course of the following description, to German Patent documents DE 34 21 253 C2 and DE 10 2005 014 237 A1.
In DE 10 2008 021 150 A1, the problem on which the present invention is also based is described in detail. This is, specifically, the ability to detect by use of an electronic control unit whether the driver of a vehicle equipped with a driver assistance system, and particularly with a driver assistance system that can automatically at least slightly steer the vehicle, is sufficiently attentive and, in particular, is capable of himself taking over the steering of the vehicle at any time. As the vehicle is moving along, the moment imparted on the steering wheel by the driver's hand will be measured, which can naturally only result in an amount unequal to zero if the driver at least touches the steering wheel. As long as the amount of the moment imparted by hand is significantly greater than zero, a sufficient driver activity can be inferred.
In the above-mentioned document, it is also described that, if no hand-imparted moment of a magnitude greater than zero is detected for only a brief time period, one does not immediately want to conclude that there is a lack of driver activity. DE 10 2008 021 150 A1 therefore describes a special method as to how, for a certain time period, the evaluation can advantageously take place with respect to the hand-imparted moment measured successively several times during that time. Naturally, other methods are also contemplated which may essentially lead to a comparable result, one of which will be briefly described together with the present invention.
However, the latter, specifically a different computing sequence or control sequence than that described in DE 10 2008 021 150 A1 is not the essential content of the present invention. It is rather the object of the present invention to further develop a method of detecting a lack of driver activity at the steering wheel of a moving motor vehicle such that the precision of the method will be further increased. As a result of the increased precision, a low number of false alarms because of an unnecessarily detected, only apparently lacking driver activity is to be set off, as well as, in cases in which there actually is no sufficient driver activity, such a lack of driver activity will be reliably detected in a timely manner.
This object is achieved by the following: should, in the case of a multiple successive falling-below a threshold value by current values of a magnitude of a moment imparted by hand, a lack of driver activity be inferred, the amount of the threshold value will be a function of the surface condition, particularly of the degree of unevenness, of the roadway traveled. Advantageous embodiments are described and claimed herein.
According to the invention, the threshold value, with which a, or the so-called, magnitude of a moment imparted by hand, which, for example or preferably, may be the amount of the measured moment imparted by hand, is compared, is not constant but depends on an essential marginal condition which is of a not negligible significance for an inference of a sufficient or lacking driver activity solely from the magnitude of the moment imparted by hand. According to the invention, the surface condition or the surface quality of the roadway should be considered, in which case the most essential influence factor when taking into account the road quality or the road condition, is the evenness or unevenness of the roadway. It was specifically recognized that, for a roadway which has the tendency to be uneven and, for example, has a plurality of cross-grooves and/or longitudinal grooves or fairly small potholes, for a determination of a sufficient driver activity, a moment imparted by hand should be required whose amount should be larger than when the vehicle is traveling on an absolutely even road, for example, formed by smooth asphalt. In the case of a good surface condition of the road that is also distinguished by a low level of unevenness, in particular, of small dimensions, a significantly lower threshold value for the magnitude of moments imparted by hand can therefore be set than in the case of a poor road condition which may be characterized, for example, by a plurality of fairly small or fairly large potholes. However, it is explicitly pointed out that the term “surface condition of the road” is not limited to the “unevenness” or “evenness” characteristic, but different threshold values may also be provided as a function of the coefficient of friction between the road and the wheels or tires of the vehicle. It is known that this coefficient of friction can be estimated with sufficient precision.
In principle, various possibilities exist for detecting the surface condition of the road. One possibility is an automatic optical analysis, for example, by analyzing camera images of the road, or by, for example, laser beam or ultrasound scanning; the latter also being included in the concept of optical analysis. However, basically further methods are also known, for the purpose of which reference is made to the documents mentioned at the outset. Thus, for example, a so-called degree of unevenness of the road can be derived from a measured vertical acceleration in the wheel suspension (for example, in the shock absorber) of the vehicle. As an alternative or in addition, it is also contemplated to derive the degree of unevenness of the road from deceleration and/or acceleration signals of at least one wheel of the vehicle, which can be obtained from the rotational wheel speeds. Within the scope of such an analysis, corresponding sensor signals can basically be subjected particularly to a low-pass filtering, and/or the signal noise can be evaluated. Independently of which method is used for detecting the road condition, the detected road condition can then be classified, and a threshold value for the magnitude of the moment imparted by hand can then be selected, which threshold value is assigned to this classification and with which the successively determined values of the magnitude of the moment imparted by hand are compared.
Such a comparison of successively determined values of the magnitude of the moment imparted by hand with the current threshold value preferably takes place for a certain time period while the vehicle is moving, so that, only if several successive measured values of the magnitude of the moment imparted by hand are below the relevant threshold value, a lack of driver activity is inferred. If, in contrast, a determined value of the magnitude of the moment imparted by hand exceeds a pertaining minimal value, preferably sufficient driver activity will be inferred and, as a result, a new series of successive measurements of the moment imparted by hand will be started, in which case the preceding measurements (before that measured value that has exceeded the minimal value) are not taken into account for the detection of a lack of activity. With the exceeding of the above-mentioned minimal value by a current value of the magnitude of the moment imparted by hand, a newly indicated time period is therefore triggered or started. Here, it should be explicitly pointed out that the mentioned time period does not have to be a firmly predefined duration but can also be set, for example, by a certain number of successive measurements of the moment imparted by hand. As far as the mentioned minimal value is concerned, the latter may be equal to the above-mentioned threshold value, or, as an alternative, differ from the latter. It may also be provided that sufficient driver activity can be inferred and therefore a new series of successive measurements of the moment imparted by hand can be started only if the magnitude of the moment imparted by hand continuously exceeds its assigned minimal value during a certain time period. This defined duration may be constant or variable; in the latter case, this duration may be a function of the surface condition of the traveled road. In addition, the above-mentioned minimal value that has to be exceeded so that a sufficient driver activity can be inferred may also be a function of the surface condition, particularly of the degree of unevenness of the traveled road.
According to an advantageous further development, the length of an above-mentioned time period within which several successive measurements of the moment imparted by hand are carried out before, in the event of a lasting falling below the threshold value, a lack of driver activity is inferred, and/or the length of the mentioned time period during which successive values of the magnitude of the moment imparted by hand have to exceed the minimal value so that a new series of successive measurements of the moment imparted by hand is started, may be depend on the driving speed of the vehicle. Normally, the time that elapses between a first and a second measurement of the moment imparted by hand will be constant and will be dependent on the clock pulse period of a CPU or the like provided in an electronic control unit in which the method according to the invention is implemented. During a longer time period, in which each measured magnitude of the moment imparted by hand has to be lower than the mentioned threshold value so that a lack of driver activity is inferred, a larger number of measured values of the moment imparted by hand is therefore considered than during a shorter time period. During a longer time period, it is therefore more probable that the driver actively operates the steering wheel during this time period and that a magnitude of the moment imparted by hand is therefore detected which exceeds the mentioned minimal value. Since, in the case of a lower driving speed of the vehicle, a brief inactivity of the driver naturally has a lower risk potential than at a higher driving speed of the vehicle, the above-mentioned time period, within which successively determined values of the magnitude of the moment imparted by hand are compared with a threshold value and in which case, when all values of the magnitude of the moment imparted by hand are below the threshold value within this time period, a lack of driver activity is inferred, may be longer at low driving speeds than at higher driving speeds of the vehicle.
For example, at a driving speed of 10 km/h, this mentioned time period may be in the order of 5 seconds and, at a driving speed of 40 km/h, may amount to only 2 seconds. A linear interpolation, for example, is also contemplated between these example values. It is also explicitly pointed out that, instead of a predefined time period, it may naturally also be predefined how many successive measured values of the moment imparted by hand or values of the magnitude of the moment imparted by hand derived therefrom have to be positively checked for a falling below the threshold value before a lack of driver activity can be inferred. Such a predefined number of measured values also may be variable, and particularly be a function of the driving speed of the vehicle and should therefore in the present case be included in the term “variable time period” or the time period dependent on the driving speed of the vehicle.
As mentioned above, the amount of the factually measured moment imparted by hand can be used as the above-mentioned magnitude of the moment imparted by hand. However, as an alternative, the use of the gradient of the moment imparted by hand, i.e. the time-related derivation of the moment imparted by hand or a use of the rate of change of the moment imparted by hand is also possible, as indicated in DE 10 2008 021 151 A1. A further increase of the precision of the method according to the invention can be achieved if, in addition to the amount of the moment imparted by hand as the first magnitude of the moment imparted by hand, the gradient of the moment imparted by hand is taken into account as a further, second magnitude of the moment imparted by hand. Only in the event that, in several successive measurements, the amount of the first magnitude of the moment imparted by hand does not exceed the first threshold value dependent on the road condition, and that the second magnitude of the moment imparted by hand does not exceed a second threshold value provided for this magnitude, will a lack of driver activity be inferred. In this case, the threshold value for the second magnitude of the moment imparted by hand may be constant or may also be a function of the road condition or of the surface condition of the traveled road.
When checking two (or more) different magnitudes of the moment imparted by hand with respect to falling below a respectively assigned threshold value within a predefined, preferably variable time period, preferably as early as when only one of these magnitudes of this moment imparted by hand exceeds its pertaining threshold value, a new series of successive measurements of the moment imparted by hand (and thereby with a new time period) will be started, so that the preceding measurements are not taken into account for the detection of a lack of driver activity. At this point, it should again be explicitly pointed out that it is not absolutely necessary to use a firmly predefined or variable time period for checking successive values of the magnitude(s) of the moment imparted by hand with respect to their threshold value. The number of successive measured values of the moment imparted by hand may rather also be defined, in which case, this predefined number may also be a function of the driving speed of the vehicle.
Other objects, advantages and novel features of the present invention will become apparent from the following detailed description of one or more preferred embodiments when considered in conjunction with the accompanying drawings.